For most charge point operators there has only ever been one revenue line: buy electricity, sell it on at a markup, and hope utilisation covers the fixed costs. That picture is starting to look incomplete. Industry analysis published this year describes flexibility as a second job the same hardware can do: getting paid by grid operators and balancing markets to shift or curtail charging load when the grid needs it.
This is not a pilot-stage idea anymore. Smart charging, demand response and vehicle-to-grid all have working commercial models in 2026, and regulation is moving them from optional toward expected. The EU's network code on demand response is expected toward 2027, which shifts flexibility participation from a nice extra into something the market assumes you can do.
Three tiers, one asset
The flexibility stack is usually described in three layers, and they all run on hardware you have already paid for:
- Smart charging. Shifting consumption into cheaper hours, using time-of-use differences in price. Runs through OCPP and ISO 15118.
- Demand response. Getting paid to curtail load during grid stress, via capacity and balancing markets. Runs through OpenADR and OSCP.
- Vehicle-to-grid. Sending energy back to the grid at the moment it is worth most, supported by ISO 15118-20 and OCPP 2.x.
The connector, the grid connection behind it and the software that controls both are already on your balance sheet. Flexibility revenue is about making that asset earn on hours when it would otherwise be sitting largely idle.
Why most operators cannot claim it yet
Getting paid for flexibility is not a contract you sign and then collect on. It is a software requirement underneath that contract. It needs real-time visibility of load per charger rather than per site. It needs the ability to actually shift or throttle a live session on an incoming signal. And it needs a way to reconcile what you were paid to curtail against what you would have earned selling that energy instead, because without that last piece you cannot tell a good flexibility offer from a bad one.
A network running a single flat tariff, set once and left alone, has none of that. There is no live read on demand per connector, no mechanism for turning an external signal into a decision at the charger, and no baseline to measure a curtailment against. Such a network cannot participate even when a local flexibility programme shows up asking for bids.
It is the same plumbing as dynamic pricing
This is the part worth sitting with. The gap that keeps an operator out of flexibility markets is the same gap that keeps them on a static retail tariff. Both need one loop: read a live signal, turn it into a decision at the connector, repeat every few minutes. Dynamic retail pricing points that loop at drivers. Flexibility points it at the grid. The counterparty changes, the machinery does not.
We run our own charging network, and we built pricing software out of what we ran into operating it. The sequencing mattered. Dynamic retail pricing had to work first: demand signals in, a priced decision out, on every connector. Once that loop existed, flexibility stopped looking like a separate integration project and started looking like the same loop with a different input. That is not a happy accident, it is what the shared requirement looks like from the inside.
We wrote earlier about flexible grid connections, where the grid operator can curtail you at your busiest moment. This is the mirror image: instead of absorbing a limit imposed on you, you offer one yourself and get paid for it. Both require the same thing, which is a network that can respond to a signal rather than one that only knows a fixed price.
What this means for planning
None of this argues for chasing a flexibility contract before the fundamentals are in place. An operator still fighting uptime, still pricing by gut feel, still without a live read on utilisation per site, has more pressing work than modelling balancing-market payments. Flexibility revenue is not a rescue for a network that is not working.
What it does argue for is treating the pricing layer as an investment that opens two doors instead of one. The retail case for demand-aware pricing already stands on its own merits. The fact that the same infrastructure is the entry ticket to flexibility revenue is a second reason to build it now rather than later, before these programmes get more crowded and the terms get less generous for latecomers.
The operators best positioned here are not necessarily the largest. They are the ones who already treat pricing as software rather than as a number set once and left alone.
Source: The EV Charging Flexibility Revenue Stack (2026), Codibly.
Frequently asked questions
What exactly are flexibility payments for a charge point operator?
They are payment for shifting or curtailing your charging load when the grid needs it, rather than payment for the kWh you sell to a driver. The stack is usually described in three layers: smart charging (moving consumption into cheaper hours), demand response (getting paid to curtail during grid stress) and vehicle-to-grid (sending energy back when it is worth most). All three run on the connector and the grid connection you have already paid for.
Why can't most charging networks participate today?
Because it is not a contract problem, it is a software problem. Participation needs real-time visibility of load per charger rather than per site, the ability to throttle a live session on an incoming signal, and a way to reconcile what you were paid to curtail against what you would have earned selling that energy. A network on a single flat tariff, set once and left alone, has none of those three.
What does this have to do with dynamic pricing?
It is the same machinery pointed at a different counterparty. Both need one loop: read a live signal, turn it into a decision at the connector, repeat every few minutes. Dynamic retail pricing aims that loop at drivers, flexibility aims it at the grid. An operator already running that loop for pricing is, by that fact, at the starting line for flexibility.
Should I be working on this right now?
Not if the fundamentals are not in place. A network still fighting uptime or still pricing by gut feel has more pressing work than modelling balancing-market payments. But it is a reason not to postpone the pricing-layer investment, since that same layer opens two doors instead of one, and European regulation on demand response is moving toward 2027.